U.S. patent application number 13/510022 was filed with the patent office on 2012-09-06 for calibration method and angle measuring method for an angle measuring device, and angle measuring device.
This patent application is currently assigned to LEICA GEOSYSTEMS AG. Invention is credited to Heinz Lippuner, Knut Siercks, Urs Vokinger.
Application Number | 20120222465 13/510022 |
Document ID | / |
Family ID | 42124635 |
Filed Date | 2012-09-06 |
United States Patent
Application |
20120222465 |
Kind Code |
A1 |
Lippuner; Heinz ; et
al. |
September 6, 2012 |
CALIBRATION METHOD AND ANGLE MEASURING METHOD FOR AN ANGLE
MEASURING DEVICE, AND ANGLE MEASURING DEVICE
Abstract
The invention relates to a calibration method that can be
carried out without a reference system for an angle measuring
device having a code carrier carrying an absolute position code,
and at least two reading heads comprising a fixed, known angle
position at an angular distance, wherein the code carrier can be
rotated relative to the reading heads, and different angle
positions of the code carrier relative to the reading heads can
thus be captured. Angle position values of the reading heads in an
angular setting are determined and angular error is determined,
which are repeated. And, a mathematical analysis method is
performed, including determining the parameters of a mathematical
function quantifying the angular error, and determining calibration
parameters as parameters of the quantifying mathematical function
or as a correction or code table derived from the parameters.
Inventors: |
Lippuner; Heinz; (Rebstein,
CH) ; Vokinger; Urs; (Au, CH) ; Siercks;
Knut; (Moerschwil, CH) |
Assignee: |
LEICA GEOSYSTEMS AG
Heerbrugg
CH
|
Family ID: |
42124635 |
Appl. No.: |
13/510022 |
Filed: |
November 25, 2010 |
PCT Filed: |
November 25, 2010 |
PCT NO: |
PCT/EP10/68259 |
371 Date: |
May 16, 2012 |
Current U.S.
Class: |
73/1.75 |
Current CPC
Class: |
G01D 5/24452
20130101 |
Class at
Publication: |
73/1.75 |
International
Class: |
G01C 25/00 20060101
G01C025/00 |
Foreign Application Data
Date |
Code |
Application Number |
Nov 26, 2009 |
EP |
09177211.1 |
Claims
1.-15. (canceled)
16. A calibration method that can be carried out without a
reference system for an angle measuring device including: a code
carrier carrying an absolute position code; and at least N.gtoreq.2
reading heads which have a fixed, known angular position having an
angular distance, in particular of more than 50 degrees, for
example of 70 to 180 degrees, preferably of 140 to 170 degrees, for
example of 150 degrees, relative to one another and in each case
detect the position code at least partially, such that an absolute
angular position value of the respective reading head relative to
the code carrier can be determined, wherein the code carrier is
rotatable relative to the reading heads and different angular
positions of the code carrier relative to the reading heads can
thus be assumed, the method comprising the following steps:
determining the angle position values of the reading heads in one
angular position, determining an angle error by comparing the
difference between the angle position values of the reading heads
with the known angular position of the reading heads with respect
to one another, repeating the steps of determining the angle
position values and the angle error for a multiplicity of varying
angular positions, and carrying out a mathematical evaluation
method including: determining parameters of a mathematical function
quantifying the angle errors, and determining calibration
parameters as the parameters of the quantifying mathematical
function or as a correction or code table which is derived from the
parameters.
17. The calibration method as claimed in claim 16, wherein the
angular positions of the reading heads are chosen in such a way
that at least one of the angular distances between respectively
adjacent reading heads differs from at least one of the remaining
N-1 angular distances.
18. The calibration method as claimed in claim 17, wherein all N
angular distances are different.
19. The calibration method as claimed in claim 16, wherein the
determination of the parameters of the quantifying mathematical
function is effected by a parameter estimation on the basis of an
optimization method.
20. The calibration method as claimed in claim 16, wherein: the
determination of the parameters of the quantifying mathematical
function is effected by a parameter estimation on the basis of an
optimization method, and the quantifying mathematical function is a
Fourier series, and represents the parameters by expansion of the
angle errors determined for the respective angular positions the
coefficients of all harmonic oscillations up to an order .gtoreq.N
of the angle error.
21. The calibration method as claimed in claim 16, wherein
repeating the steps of determining the angular position values and
the angle error takes place in a measurement range of the angle
measuring system, which measurement range is subdivided into
subranges, in such a way that specific subranges are covered by the
varying angular positions with a defined density.
22. The calibration method as claimed in claim 21, wherein the
angular positions are distributed approximately uniformly over said
specific subranges and the angular position values are detected
simultaneously at the reading heads.
23. The calibration method as claimed in claim 16, wherein: in
addition a determination of misalignments by monitoring the
calibration parameters for change; and/or a derivation of an
estimated value for the accuracy of the angle sensor on the basis
of the calibration parameters are/is carried out.
24. The calibration method as claimed claim 16, wherein the angle
position value and/or the angle error are/is determined multiply in
the same angular position and these are averaged before the
mathematical evaluation method is carried out.
25. The calibration method as claimed claim 16, wherein the angle
position value and/or the angle error are/is determined multiply in
the same angular position and these are averaged before the
mathematical evaluation method is carried out by means of a
compensatory calculation according to the least squares method, or
the calibration method is performed multiply and an optimization
method is carried out for the sets of coefficients obtained in the
process.
26. The calibration method as claimed in claim 16, wherein the
angular position of the reading heads is determined on the basis of
a difference between the angle positions in an arbitrary reference
angular position.
27. The calibration method as claimed in claim 16, wherein the
angular position of the reading heads is determined by averaging
the differences over a plurality of arbitrary reference angular
positions.
28. The calibration method as claimed in claim 16, wherein: the
angular position of the reading heads is chosen in such a way that
the dominant angle errors, which supply a significant error
proportion, including: errors caused by eccentricity, errors in the
code division of the code disk, and/or errors owing to changing
ambient influences, are contained in the qualifying mathematical
function with an error proportion of at least 25%.
29. The calibration method as claimed in claim 16, wherein the
angle positions are determined with reading heads arranged
symmetrically in each case in pairs.
30. The calibration method as claimed in claim 16, wherein the
reading heads are embodied as optical, electro-optical, magnetic,
or capacitive reading heads.
31. An angle measuring method for determining an angle measurement
value with the aid of an angle measuring device, wherein the angle
measuring device includes: a code carrier carrying an absolute
position code, and at least N.gtoreq.2 reading heads which have a
fixed, known angular position having an angular distance, in
particular of more than 50 degrees, for example of 70 to 180
degrees, preferably of 140 to 170 degrees, for example of 150
degrees, relative to one another, in particular wherein the angular
positions of the reading heads are chosen in such a way that at
least one of the angular distances between respectively adjacent
reading heads differs from at least one of the remaining N-1
angular distances, and in each case detect the position code at
least partially, such that an absolute angular position value of
the respective reading head relative to the code carrier can be
determined, wherein the code carrier is rotatable relative to the
reading heads and different angular positions of the code carrier
relative to the reading heads can thus be assumed, the method
comprising the following steps: determining an angle position value
by means of at least one of the reading heads, which detects at
least one part of an absolute position code of a code carrier;
determining the angle measurement value by correcting the angle
position value on the basis of calibration parameters determined in
the context of a calibration method as claimed in claim 16; and
providing the angle measurement value.
32. The angle measuring method as claimed in claim 31, wherein in
addition to the calibration method as claimed in claim 16 is
carried out: during assembly of the angle sensor, after the
mounting of angle sensor in a device, upon each occasion when the
device is switched on, in cyclic or acyclic time intervals, in a
manner initiated by the user, when there is a change in the ambient
conditions, after impact or shock stresses, at regular service
intervals and/or continuously during measurement operation, taking
account of calibration parameters which were determined during
measurement operation and/or were determined during an earlier
calibration and stored.
33. Angle measuring method as claimed in claim 32, wherein after
the calibration method has been performed, one reading head or a
plurality of reading heads is/are deactivated or removed.
34. An angle measuring device comprising: a code carrier carrying
an absolute position code, at least N.gtoreq.2 reading heads which
have a fixed, known angular position having an angular distance, in
particular of more than 50 degrees, for example of 70 to 180
degrees, preferably of 140 to 170 degrees, for example of 150
degrees, relative to one another, in particular wherein the angular
positions of the reading heads are chosen in such a way that at
least one of the angular distances between respectively adjacent
reading heads differs from at least one of the remaining N-1
angular distances, and in each case are suitable for detecting the
position code at least partially, such that an absolute angular
position value of the respective reading head relative to the code
carrier can be determined, and wherein the code carrier is
rotatable relative to the reading heads and different angular
positions of the code carrier relative to the reading heads can
thus be assumed, and an evaluation unit for carrying out the
following steps of a calibration method including: determining an
angle error by comparing the difference between the angle position
values of the reading heads with the known angular position of the
reading heads with respect to one another; repeating the steps of
determining the angle position values and the angle error for a
multiplicity of varying angular positions; carrying out a
mathematical evaluation method comprising determining the
parameters of a mathematical function quantifying the angle errors;
determining calibration parameters as the parameters of the
quantifying mathematical function or as a correction or code table
which is derived from the parameters; and/or at least the following
steps of the angle measuring method of claim 31 including:
determining the angle measurement value by correcting the angle
position value on the basis of calibration parameters determined in
the context of the calibration method of claim 16; providing the
angle measurement value.
35. A computer program product stored on a machine-readable
carrier, comprising program code for carrying out the following
steps of the calibration method as claimed in claim 16: determining
the angle error by comparing the difference between the angle
position values of the reading heads with the known angular
position of the reading heads with respect to one another, wherein
the angular positions of the reading heads are chosen in such a way
that at least one of the angular distances between respectively
adjacent reading heads differs from at least one of the remaining
N-1 angular distances; repeating the steps of determining the angle
position values and the angle error for a multiplicity of varying
angular positions; carrying out a mathematical evaluation method
comprising: determining the parameters of a mathematical function
quantifying the angle errors; determining calibration parameters as
the parameters of the quantifying mathematical function or as a
correction or code table which is derived from the parameters;
and/or at least the following steps of the angle measuring method
as claimed in claim 31: determining the angle measurement value by
correcting the angle position value on the basis of calibration
parameters determined in the context of the calibration method as
claimed in claim 16; providing the angle measurement value.
Description
[0001] The invention relates to a calibration method and an angle
measuring method for an angle measuring device according to the
preamble of claims 1 and 11, respectively, and to a computer
program product according to the preamble of claim 15 for carrying
out the methods and an angle measuring device according to the
preamble of claim 14 and with the aid of which a multiplicity of
systematic measurement errors can be compensated for.
[0002] Angle measuring systems have been known in a wide variety of
embodiments for many years. They are employed in mechanical and
plant engineering and likewise in a wide variety of measuring
devices, for example in geodetic devices, coordinate measuring
devices or robots. When they are employed, the objective that
arises is to determine the angle deviation or corresponding
variables derived therefrom, such as velocity or acceleration, for
example, between two subsystems rotatable relative to one another
in one degree of freedom.
[0003] By way of example, such angle measuring systems are used in
coordinate measuring devices such as e.g. articulated arms for
determining the angle positions in the individual articulations,
from which the spatial position of a sensing element fitted to the
end of the arm is subsequently calculated.
[0004] Such angle measuring devices are also incorporated in
geodetic measuring devices, such as a theodolite, for example,
which are used to carry out a wide variety of measurement tasks,
such as, for example, determining horizontal and vertical
angles.
[0005] Such angle measuring systems can likewise be found in plants
and machines for detecting the positions of machine components such
as drives, pivoting heads, throttle valves, rotary tables, and the
like. The positions detected in this case can be utilized as
position values for measurement purposes, or else for a positioning
of components by a drive with a position control loop.
[0006] An angle measuring system is constructed from two subsystems
movable relative to one another in one degree of freedom. The first
subsystem carries a position code, which is detected wholly or
partially by a reading head fitted to the second subsystem. By
evaluating the signals of the reading head, therefore, an
evaluation unit can determine the position of the two subsystems
with respect to one another. If the position code is an absolute
code, an unambiguous angle position value of the two subsystems
with respect to one another can be determined at least in sections.
In this case, use is often made of a code table for converting the
position code into an angle position value.
[0007] A large number of the angle measuring systems available
nowadays comprise a plurality of reading heads for determining the
angle position values with increased measurement accuracy, for
example by reducing the non-systematic errors, such as signal
noise, for example, by averaging the individual angle position
values. In other applications, a plurality of reading heads can
also be utilized for avoiding erroneous measurement values by
redundancy.
[0008] Owing to the systematic nature of a large number of the
errors that occur, the latter are not sufficiently corrected by
such averaging. In particular errors which are harmonic with
respect to a full revolution, especially low-order harmonic errors
such as arise, for example, in the case of rotary encoders as a
result of eccentricity, bearing errors, code division errors, etc.,
in practice often make up a large proportion of the total error.
Therefore, it is of particular interest to detect them and
correspondingly correct the measurement value. A wide variety of
possibilities for calibrating the angle measuring device are known
for this purpose.
[0009] The calibration of highly accurate angle measuring devices
requires extremely high precision calibration apparatuses whose
accuracy has to be significantly higher than that of the test
specimens to be calibrated. Therefore, such apparatuses are rather
complicated to produce and hence cost-intensive. The calibration
process is often also associated with very high expenditure of time
and labor. It is therefore endeavored to automate this process as
far as possible and to dispense with expensive apparatuses.
[0010] The publication EP 0 440 833 B1 therefore describes an angle
measuring device in which systematic errors are detected and
corrected with the aid of a plurality of reading heads, on the
basis of a discrete recursive calculation. In this case, for
complete correction of the errors over the entire measurement
range, owing to the incremental encoding, either a large number of
reading heads are required or it is necessary to employ an external
reference system which predefines known positions externally. In
practice, such a recursive calculation, owing to error propagation
and the ever present noise components, also soon encounters limits
with regard to the achievable accuracy.
[0011] DE 11 2006 003 663 T5 discloses the expansion of the angle
errors in a Fourier series, which is particularly suitable for
describing and correcting harmonic errors of a rotary encoder over
the entire measurement range of 360.degree.. Higher-order harmonic
angle errors can be determined accurately with the individual,
divided reading head used in that case. Since the small angle
covered by the divided reading head is insufficient in the case of
the Fourier analysis used for accurately determining low harmonics,
it is only by means of a predetermined external positioning of the
subsystems with respect to one another that these low-order errors
are also made accurately determinable. By way of example, for
determining the first harmonic with a Fourier decomposition,
measurements at as far as possible opposite positions are
advantageous, since said first harmonic can thus be determined to a
high proportion and therefore also with a correspondingly good
signal-to-noise ratio (see equation 1 below). For this purpose,
however, an external reference is necessary in said disclosure,
which external reference positions the encoder in predetermined
positions in order also to make the low harmonic determinable with
sufficient accuracy.
[0012] The document EP 1 944 582 A1 discloses a method which, for
determining at least one influencing variable that influences the
eccentricity in an angle measuring device with a detector
arrangement composed of four optical detector elements, a rotatable
rotary body with a multiplicity of pattern elements arranged around
a pattern center, which are at least partially imaged onto the
detector arrangement, resolves the positions and determines the
eccentricity of the pattern center relative to a detector center of
the detector arrangement. The arrangement of the four detectors
used in that case is symmetrical, with a uniform division of the
detectors along the circumference of the rotary body, as a result
of which, for example, the fourth harmonic cannot be
determined.
[0013] The document DE 103 57 602 A1 likewise describes an angle
sensor in which, during the calibration process, at externally
predetermined angle positions, the difference between the
measurement value and the externally predetermined angle position
is determined and expanded in a Fourier series. The series
coefficients are stored and said coefficients, once they have been
stored, are used from then on for correcting the measurement
values. In that case, too, calibration is possible only with the
aid of an external reference.
[0014] The article "Self Calibrating Rotary Encoder" by Tsukasa
Watanbe et al., which was published in the Journal of Physics,
series 13, 2005, on pages 240-245, describes a self-calibrating
rotary encoder operating with a plurality of reading heads arranged
equidistantly on the circumference.
[0015] In the journal "Measurement Science and Technology" issue
19, 2008, R. Probst describes in the article "Self-calibration of
divided circles on the basis of a prime factor algorithm"
self-calibration of divided circles using a prime factor
decomposition for a discrete Fourier transformation.
[0016] What is disadvantageous about the known methods is that the
error correction, in particular of the low-order harmonic errors,
can be determined with sufficient accuracy only with the aid of an
externally predefined positioning of the angle measuring system.
Consequently, a calibration for reducing these errors is possible
only with external aids. This considerably complicates the
calibration, particularly after the angle measuring device has been
incorporated into a device.
[0017] Furthermore, there is also no simple possibility for renewed
calibration of the finished device in order, for example, to be
able to react to changing ambient conditions or to potential
sources of error such as shock and impact stresses, since a highly
accurate external positioning of the sensor incorporated in the
device is often difficult or even impossible.
[0018] Against this background, therefore, the general object that
arises is to improve the angle accuracy of an angle measuring
system by compensating for errors present as a result of a
calibration.
[0019] One specific object, therefore, is to provide a calibration
method having improved error compensation.
[0020] In particular, it is an object to provide a calibration
method which operates without a reference system and with which a
calibration can be carried out without any external aids and angle
predefinition. An angle measuring system equipped with an for such
a calibration method therefore affords the possibility of being
calibrated by itself, at any time without the presence of a complex
calibration unit. Such a calibration without a reference system and
without external aids is also designated hereinafter as
self-calibration.
[0021] In this case, one specific object of the invention is also
to achieve error correction with high accuracy, particularly for
the most dominant errors of the angle measuring system, which,
particularly in the case of angle measuring devices, often occur in
the form of a low-order error that is harmonic with respect to a
full revolution.
[0022] In this case, furthermore, the fewest possible reading heads
should also be used, in order to keep down the costs, complexity
and error susceptibility of the angle measuring system.
[0023] The energy requirement of an angle measuring system,
particularly in the case of mobile use, should also be kept as low
as possible, in order, for example, to be able to ensure a long
operating time of a battery-operated geodetic measuring device.
Therefore, it is a further object to use the fewest possible
reading heads and an as far as possible non-computationally
intensive correction algorithm. Fewer reading heads and a simple
correction method additionally enable a higher evaluation speed of
the measuring system and thus the advantage of a shorter
measurement time for the user.
[0024] These objects are achieved by the realization of the
characterizing features of the independent claims. Features that
develop the invention in an alternative or advantageous manner can
be gathered from the dependent patent claims.
[0025] The angle measuring device according to the invention
achieves said objects by means of the arrangement according to the
invention of a plurality of reading heads determining an absolute
position and an associated method for determining error
coefficients and a method for correcting the measurement value on
the basis of the error coefficients.
[0026] In this case, by means of a corresponding arrangement of the
reading heads and an absolute encoding, an accurate determination
of the harmonic errors, particularly of the low-order harmonic
errors that are often dominant, is made possible solely by the
angle measuring system itself. These errors determined can
subsequently be taken into account on the basis of the described
correction method when determining a measurement value and a
corrected measurement value can thus be output for further
processing.
[0027] Harmonic errors can originate, for example, from
eccentricities during mounting, poor storage, oblique mounting,
division errors of the code disk, wobbling of the code disk,
thermal or mechanical warpage, misalignment of the code disk, aging
phenomena or other circumstances.
[0028] In practice, low-order harmonics, for example up to the 10th
or 15th order, are often proportionately dominant in this case.
Therefore, it is also deemed necessary to detect particularly these
harmonics as accurately as possible, since they make up the largest
proportion of the total error. In the case of angle sensors, in
this case especially the dominant angle errors as harmonics up to
at least the eighth order supply a significant error proportion,
and should therefore be contained in the qualifying mathematical
function or series with high accuracy, at least with a
signal-to-noise ratio of at least 25%.
[0029] The angle measuring system used is composed of at least two
subsystems moveable relative to one another in one degree of
freedom. In this case, the first of said subsystems carries a
position code, and the second subsystem is equipped with at least
two reading heads for detecting the position code or a part
thereof. Alternatively, a plurality of subsystems connected to one
another and each having one or more reading heads can also be
present. In any case, the arrangement or the design of the reading
heads must be suitable for detecting the position code at at least
two different angular positions.
[0030] The at least two reading heads are fitted relatively to one
another at a fixed, known angular distance. The angular positions
are chosen in such a way that at least one of the angular distances
between two adjacent reading heads differs from the remaining
angular distances, in particular wherein the reading heads can have
angular distances of more 50 degrees relative to one another.
[0031] A reading head having an elongate or two-dimensional sensor
can likewise be used if, with the latter, the position code can be
read out according to the invention in at least two angular
positions which are complained relative to one another in
particular by an angle of at least 50 degrees. As a result, the two
or more reading heads described here can for example also be
realized by an individual physical reading head for the purposes of
the invention.
[0032] The position code is an absolute position code, from which
it is possible to determine an absolute angle position of the
reading head at the circumference of the code carrier. Alongside
the embodiment as a code disk, rotary encoders having a code
carrier are also known in which the angle division is fitted on
wound tapes and the position code is read from said tapes.
[0033] The encoding of the code carrier can be embodied for example
in the form of one code track having serial absolute encoding (e.g.
having a maximum sequence) or a plurality of code tracks on the
basis of which the position can be determined absolutely (e.g. with
more parallel code tracks which represent a binary position
encoding). It is also possible merely to use an only partly
implemented absolute encoding which is supplemented by an
intervening relative encoding.
[0034] The angular positions at which the reading heads detect the
position code directly influence the determinability of the
harmonics. Depending on the angular distance between the reading
heads--designated hereinafter by .DELTA.--the weighting factors of
the harmonics result according to the following formula:
a h = 2 A h sin ( h .DELTA. 2 ) , ( 0.1 ) ##EQU00001##
[0035] In the case of more than two reading heads, it is possible
in this case, of course, to utilize any desired permutations of the
reading heads for determining the differences.
[0036] As already described above, by way of example, the first
harmonic can be determined particularly advantageously by a
difference between the angular positions of 180.degree.. However,
in the case of such an arrangement, by way of example, the second
harmonic cannot be determined at all.
[0037] At small angles of less than 50 degrees between the reading
heads, the first- and second-order harmonics can only be determined
with low weighting factors, usually below 50%. Since a multiplicity
of systematic errors occur precisely with this low periodicity,
however, they can be determined without a reference system with
reading head distances of less than 50 degrees only to a very
limited extend with the accuracy required for a precision
measurement.
[0038] From the angle position values determined by evaluation of
the reading heads at the positions, by means of a series expansion,
in particular by means of a Fourier series expansion of the
form:
.THETA. ~ k = .PHI. + .OMEGA. k + k ( .THETA. k ) = .THETA. k +
.eta. + h = 1 H A h , k cos ( h .THETA. k + .PHI. h , k ) ( 0.2 )
##EQU00002##
the angle errors of the angle position values are represented as
harmonic oscillations in the form of coefficients of said
series.
[0039] The formulation describes the angle error at a reading head
k of an angle measuring system having K heads which are fixedly
mounted with respect to a reference angle, on the ideal reference
system, at the angular positions .OMEGA..sub.k, and .PHI. is the
angle of the ideal reference system. The deviation
( .THETA. k ) = .eta. k + h = 1 H A h , k cos ( h .THETA. k + .PHI.
h , k ) ( 0.3 ) ##EQU00003##
of the angular position value .THETA. from the ideal desired angle
of the reference system
.THETA..sub.k=.PHI.+.OMEGA..sub.k (0.4)
is described in equation (0.2) by the sum of cosine terms and/or
sine terms in accordance with a Fourier expansion and a noise term
.eta..sub.k. If the angular position between two arbitrary reading
heads
.THETA. ~ k - .THETA. ~ f = .THETA. k + .eta. k - ( .THETA. f +
.eta. f ) + h = 1 H A h , k ( cos ( h .THETA. k + .PHI. h , k ) -
cos ( h .THETA. f + .PHI. h , f ) ) ( 0.5 ) ##EQU00004##
is then calculated, the coefficients A.sub.k,h of the harmonics can
be determined by an optimization method. For this purpose,
.THETA..sub.k-.THETA..sub.f=a.sub.k,0-a.sub.f,0 (0.6)
as zeroth-order harmonic and
.THETA..apprxeq..THETA.. (0.7)
are introduced into equation (0.5); the latter on account of the
small amplitude .epsilon. to be expected.
[0040] Consequently, the following applies to equation (0.5)
k , l - .eta. k - .eta. f + h = 0 H A h , k ( cos ( h .THETA. ~ k +
.PHI. h , k ) - cos ( h .THETA. ~ f + .PHI. h , f ) ) ( 0.8 )
##EQU00005##
By means of a compensatory calculation, it is possible to determine
the values of a.sub.h,k, .phi..sub.h,k and .phi..sub.h,f and thus
to reconstruct the function .epsilon.. Therefore, the angle errors
of all the angular positions can be estimated using two angle
heads. The calculation of the at least first eight or eleven
harmonics normally suffices to achieve a high angle accuracy. The
method can be extended to an arbitrary number of angle heads and
the estimation is thereby also improved.
[0041] The angle heads can be mounted in a targeted manner for the
measurement of specific harmonics (.OMEGA..sub.k). Thus, e.g.
.DELTA..sub.k,f=.OMEGA..sub.k-.OMEGA..sub.f=45 degrees (0.9)
is optimal for the measurement of the fourth harmonic. Other
harmonics can also be determined with this angular position of the
reading heads, but with a poorer signal-to-noise ratio than in the
case of an optimal angular position.
[0042] In principle, the following relationship holds true:
a h = 2 A h sin ( h .DELTA. 2 ) , ( 0.10 ) ##EQU00006##
where A.sub.h is the real amplitude of the h-th harmonic, a.sub.h
is the measured amplitude of the h-th harmonic and .DELTA. is the
angle formed between the reading heads under consideration.
[0043] By way of example, the fourth harmonic is measured at an
angular position of 45 degrees with the amplitude a.sub.4-2A.sub.4.
However, the eighth harmonic cannot be detected at all with this
arrangement, since a.sub.8=0. As a further example, with this
angular position of the reading heads, the sixth harmonic can be
determined with an amplitude of a.sub.6=1.41A.sub.6.
[0044] The radial movement of the code disk is fully incorporated
in the calculation of the harmonics. This can lead, in the case of
a poor bearing, to high noise components in the angle position
values of the individual heads (higher value.eta.). If, with
respect to each angle head, a further angle head lying
diametrically opposite is mounted and the average values of these
head pairs in used for .THETA..sub.k, then the radial movements
cancel one another out and no longer appear as noise in the
evaluation of equation (0.8).
[0045] On advantage of the method is also that the error function
.epsilon. is continuous rather than discrete as is often the case
in the prior art.
[0046] A set of coefficients determined in this way is stored. This
set of coefficients can be used for a continuous online correction
in the form of a calculation of a corrected measurement value on
the basis of the coefficients during the measurement process.
Alternatively, with the set of coefficients, a code table of the
angle measuring device can also be corrected and it can be stored
or a separate correction table can be created and stored. The two
alternatives afford the advantage of a lower computational
complexity during the measurement, but require more memory
space.
[0047] In order to carry out a complete calibration, the angle
measuring device merely has to be moved along its measurement
range. In contrast to the prior art, it is not necessary to move to
predetermined positions. It furthermore suffices to move the angle
sensor for calibration by less than one full revolution. The
minimum distance to be traveled for calibration results from the
angular positions of the reading heads.
[0048] Further movement, beyond the minimum distance necessary for
calibration, leads to redundant data. With the aid of such
redundant data, it is possible, for example by means of averaging
or a compensatory calculation, in the case of the values of the
angle errors or series coefficients, additionally also to reduce
non-systematic components and thus to further improve the accuracy
of calibration.
[0049] On the basis of the set of coefficients determined, the
measurement value is subsequently corrected during a measurement,
and the harmonic errors mapped by the coefficients can thus be
compensated for. Alternatively, it is also possible to modify a
code table of the absolute value encoder on the basis of the
coefficients, in order to directly compensate for the errors there
and to reduce the calculation complexity during the determination
of measurement values. The coefficients or the modified code table
are/is stored either in volatile or in nonvolatile fashion by the
angle sensor or by the evaluation unit thereof. Volatile storage
requires a renewed calibration upon every switch-on or reset of the
system, but this can be carried out at any time even by a layperson
on account of the calibration method that is simple to perform.
[0050] Since no external aids and references are required for a
calibration according to the invention the calibration can be
carried out at any time and at any place, for example: [0051]
during the assembly of the angle measuring system, [0052] when the
angle measuring system is incorporated into a device, [0053] once
when the device is put into operation, [0054] upon each occasion
when the device is switched on, [0055] at predefined time
intervals, [0056] when there is a change in the ambient conditions,
[0057] after mechanical impacts, falls or vibrations, [0058]
continuously during measurement operation taking account of
calibration parameters which were determined during measurement
operation and/or calibration parameters which were determined and
stored during an earlier calibration.
[0059] Wherein in particular a calibration directly before
measurement in the field is particularly advantageous in this case,
since the current state and the ambient conditions of the measuring
device are thus taken into account.
[0060] On the basis of the calibration parameters, it is also
possible to determine an estimated value for the accuracy of the
measuring system in order to qualitatively assess the measuring
system, or the measuring system can assess itself with regard to
the accuracy achievable by it and provide this information for
further processing. Moreover, the calibration parameters can be
monitored by the device and a warning or an indication of a
possible lack of accuracy, necessary maintenance or other problems
can be issued if an unexpectedly great change in the coefficients
relative to earlier values occurs--for example caused by a
misalignment of a reading head or of the code carrier.
[0061] During the measurement, it is also possible, as necessary,
for one or more measuring heads to be deactivated or even removed.
After the determination of the coefficients or the corrected code
table, one reading head is sufficient for determining a value which
is corrected on the basis of the calibration. However, a plurality
of measuring heads enable here, too, an advantageous interpolation
or averaging for further increasing the measuring accuracy.
[0062] For the self-calibration according to the invention, no
special requirement is made of the accuracy of the known angular
positions of the reading heads with respect to one another, since
this can be determined in the course of the calibration of the
angle measuring system on the basis of the difference between the
absolute angle position values of the reading heads in an arbitrary
angular position as reference angular position. During this
determination, it is also possible, for example, to employ
averaging of a plurality of read values, also in different angular
positions, for the purpose of reducing errors. The angular distance
between the reading heads--or the known angular positions of the
reading heads--may therefore indeed only be known approximately, if
necessary also not at all, before the calibration and may only be
determined in the course of the calibration method. Consequently,
the term "known" also concomitantly includes the implicit knowledge
on account of the determinability.
[0063] However, the angular positions of the reading heads have to
remain unchanged during the determination of the set of
coefficients, where reading heads which are fixedly mounted in a
device can be taken as a basis to a great extent. On account of the
closed angle circle to be regarded as closed, an angular distance
between two reading heads with more than 180 degrees can also be
described by means of the complementary angle. By way of example,
an angular distance of more 180 degrees in the embodiments
described here is measured in the other direction and,
consequently, the maximum possible angular distance can also be
regarded as 180 degrees and not 360 degrees.
[0064] A further advantage of the method according to the invention
is that, by virtue of the fixedly positioned reading heads, a
stabler angular position can be achieved than is possible in the
case of an external positioning of the angle sensor, since a
positioning always has both absolute positioning errors and
reproducibility errors in a manner governed by the system.
[0065] The method according to the invention and the apparatus
according to the invention are described in greater detail below
purely by way of example on the basis of concrete exemplary
embodiments illustrated schematically in the drawings, further
advantages of the invention also being discussed. More
specifically:
[0066] FIG. 1a shows an embodiment of an angle measuring system
according to the invention with two reading heads arranged at an
angle .DELTA. of 150.degree. with respect to one another,
[0067] FIG. 1b shows by way of example a tabular representation of
the weighting factors with which harmonics can be determined with
the arrangement illustrated in FIG 1a,
[0068] FIG. 2 shows an embodiment of an angle measuring system
according to the invention with two reading head pairs respectively
arranged diametrically,
[0069] FIG. 3a shows an embodiment of an angle measuring system
according to the invention with two reading heads respectively
arranged opposite, and a further, individual reading head,
[0070] FIG. 3b shows an embodiment of an angle measuring system
according to the invention with two reading heads arranged
diametrically, and two further reading heads in different angular
positions, and
[0071] FIG. 4 shows a flowchart of the calibration method and of
the angle measuring method,
[0072] FIG. 5 shows a flowchart of the calibration method and of
the angle measuring method in the case of continuous
calibration,
[0073] FIG. 6a shows by way of example the use of a plurality of
angle measuring systems according to the invention in an
articulated arm, and
[0074] FIG. 6b shows by way of example the use of two angle
measuring systems according to the invention in a geodetic
measuring device.
[0075] FIG. 7 shows by way of example a representation of the
errors of an angle measuring device as a Fourier series,
[0076] FIG. 8 shows an exemplary embodiment of an angle measuring
system according to the invention with optoelectronic reading
heads.
[0077] A possible embodiment of the angle measuring system
according to the invention can be seen in a schematic illustration
in FIG. 1a.
[0078] There becomes an angle measuring system 7 comprising [0079]
a code carrier 2 carrying an absolute position code, and [0080] at
least two reading heads 1a, 1b, which have a fixed, known angular
position 4 having an angular distance, in particular of more than
50 degrees, for example of 70 to 180 degrees, preferably of 140 to
170 degrees, for example of 150 degrees, relative to one another
and in each case suitable for detecting the position code at least
partially, such that an absolute angular position value of the
respective reading head 1a, 1b relative to the code carrier 2 can
be determined, and wherein the code carrier 2 is rotatable relative
to the reading heads 1a, 1b, and different angular positions 3 of
the code carrier 2 relative to the reading heads 1a, 1b can thus be
assumed, and [0081] an evaluation unit for carrying out the
calibration method according to the invention and/or the angle
measuring method according to the invention, which is not
illustrated.
[0082] Two reading heads 1a and 1b for detecting the position code
situated on the code carrier 2 are illustrated. They are arranged,
by way of example, in a manner spaced apart from one another in an
angular position 4 of 150.degree.. An evaluation unit, not
illustrated here, determines an angle position value in each case
on the basis of the position code at least partially detected by
the reading head, 1a, 1b, said angle position value being dependent
on the position of the code carrier 2 relative to the reading head
1a, 1b. As indicated by the arrow 3 for illustration purposes, the
code disk 2 is mounted such that it is rotatable in one degree of
freedom relative to the reading heads 1a, 1b.
[0083] In this case, the reading heads 1a, 1b are embodied such
that their angle position value relative to the code carrier 2 can
be determined absolutely within one revolution or part of a
revolution. In this exemplary embodiment, the angular position 4 of
the reading heads 1a and 1b with respect to one another is
150.degree., as indicated. Consequently, in the ideal, error-free
case, the difference between the angle position values of the
reading heads 1a and 1b also makes up 150.degree.. In practice, an
angle error can be determined on the basis of a comparison of the
difference between the angle position values with the angular
position.
[0084] By rotating the code disk 2 relative to the reading heads
1a, 1b the angular position is varied. In this case, the angle
error is determined for a multiplicity of angular positions. This
is preferably done with approximately uniform distribution over the
angular positions of the entire measurement range. This would
ideally be carried out for every angular position at which a
different angle position value can be determined, but in practice
this can be realized only with difficulty and is not required
either for the method. It suffices merely to cover subranges of the
measurement range with a defined density in order to carry out the
calibration method.
[0085] The angle errors obtained in this case are expanded e.g. in
a mathematical series. By way of example a Fourier series is
suitable for this since its coefficients represent harmonics of a
fundamental period and these are therefore highly suitable for
mapping errors of a rotary encoder which are periodic with respect
to a revolution.
[0086] As an alternative to a Fourier series, which will be
discussed in more specific detail by way of example for elucidating
the method in the further description, other mathematical
functions, series and models can likewise be employed for
qualifying the angle errors. The parameters of this general
function then correspond in terms of their function to the
coefficients (presented in the description) of the harmonics of the
Fourier series and can be determined for example by one of the many
known parameter estimation methods or an optimization
calculation.
[0087] In this case, the angular positions of the reading heads are
chosen in such a way that at least one of the angular distances
between the respectively adjacent reading heads differs from at
least one, in particular from all, of the angular distances between
the remaining/further reading heads, as a result of which, in
particular, no totally symmetrical arrangement of the reading heads
along the circumference of the code carrier is provided.
Consequently, no rotationally symmetrical arrangement is involved
since the distances between the reading heads are unequal. The
reading heads do not have identical interspaces along the code
carrier.
[0088] According to the invention, by way of example, in the case
of N reading heads, at least N-1 distances between the reading
heads differ from one another, as a result of which a
determinability of harmonics up to an order .gtoreq.N is made
possible.
[0089] In other words, in accordance with the present invention,
none of the angular distances should correspond to another angular
distance, that is to say that the complaints of the reading heads
are different.
[0090] Two identical angular distances between reading heads can
admittedly make possible, as a result of the determinability of
redundant information, an improvement in the accuracy, for example
as a result of interpolation, which does not preclude the
application thereof per se, but reading heads spaced apart
identically nevertheless generally do not contribute to the
determinability of additional harmonic errors in accordance with
the invention.
[0091] By way of example, it may be endeavored to make all the
distances between the reading heads different and to choose the
latter in such a way that the weighting factors resulting from the
angular distances in accordance with the formula 0.10 turn out to
be as large as possible for all harmonics to be determined and
particularly in the joint consideration of the harmonics to be
determined over all angular distances do not turn out to be zero
for any harmonic for all angular distances that can be evaluated.
Thus, by way of example, in the case of an expected uniform
distribution of the errors over the harmonics to be determined over
all angular distances, a corresponding uniform distribution of the
sum of the weighting factors per harmonic to be determined may be
striven for. The mathematical principles for calculating a
corresponding angular distance division for a specifically desired
combination of number of reading heads, number of harmonics to be
determined and the weighting factors thereof (e.g. on account of an
expected error distribution) are sufficiently known from the
literature.
[0092] FIG. 1b illustrates the weighting factors of harmonics in
the case of a Fourier series for the arrangement of the reading
heads with an angular position of 150.degree. from FIG. 1a in
tabular form.
[0093] In this case, the first column h indicates the order of the
harmonics and the column a/A (150.degree.) indicates the associated
weighting factor of these harmonics in the range of 0% to 100%. In
this case, a weighting factor of 100% stands for the complete
detectability of these harmonics and 0% stands for a harmonic that
cannot be detected with this arrangement. These weighting factors
crucially influence the achievable signal-to-noise ratio (SNR) when
determining the harmonics. Consequently, the weighting factors also
determine the maximum achievable accuracy of the coefficient values
of the harmonics.
[0094] It can be seen from the table that harmonics up to the
eleventh order can be determined throughout with an angular
position of 150.degree.. In this case, the 1st, 4th, 6th, 8th and
11th harmonics can be determined with a comparatively good SNR by
virtue of their high weighting factor. The 12th harmonic cannot be
determined with this arrangement since its weighting factor is
0%.
[0095] Consequently, at 150 degrees, the first harmonic up to the
eleventh harmonic can be determined continuously, but with
different signal-to-noise ratio.
[0096] The angle of 150 degrees was chosen here by way of example.
It goes without saying that any other angles can be chosen, for
example in order to be able to detect harmonics of a desired order
with the highest possible quality if such a requirement is known.
In the case of a very small angular distance between the reading
heads of less than 50 degrees, however, the weighting factors of
the low harmonics, in particular of the first harmonic, are very
small. Therefore, with small angular positions of less than 50
degrees, typical errors of rotary encoders such as [0097]
unbalance, [0098] eccentricity, [0099] wobbling errors, [0100]
bearing errors, [0101] thermal or mechanical warpage of the code
carrier can only be detected with restricted accuracy.
[0102] The principle can also be extended to more than two reading
heads and to permutations thereof, in order to be able to detect
further harmonics or some harmonics multiply and thus to be able to
determine the angle errors even better.
[0103] In this case, the coefficients obtained in the series
expansion represent the error in the form of harmonic components
over the periodic measurement range of one revolution. A
measurement value of the angle measuring system can be corrected on
the basis of the coefficients and the angle position value of the
reading head. As an alternative thereto, it is also possible, with
the coefficients, directly to modify a code table used during the
absolute value determination or to create a correction table with
respect to the code table and thus to dispense with correction
calculations during measurement operation.
[0104] For complete calibration, the angle sensor has to be rotated
by a specific, minimal angle dependent on the angle positions of
the reading heads. As a result, the coefficients of the harmonics
can be detected. The redundant data determined by rotation going
beyond that, which data, for example by means of a compensatory
calculation, can be utilized also for reducing non-systematic
errors and for further improving the accuracy of the
calibration.
[0105] After calibration has been effected, during a measurement it
is also possible for one or a plurality of the reading heads to be
deactivated or even removed. As a result, by way of example, the
current consumption can be reduced and/or the evaluation speed
increased.
[0106] The calibration of the angle measuring system can be
effected at a wide variety of points in time, in particular: [0107]
during the assembly of the angle measuring system, [0108] after the
angle measuring system has been mounted in a device, [0109]
continuously during measurement operation, [0110] when the device
is first switched on, [0111] upon each occasion when the device is
switched on, [0112] in cyclic or acyclic time intervals, [0113]
when there is a change in the ambient conditions, [0114] after
impact or shock stresses, [0115] at regular service intervals,
[0116] continuously during measurement operation taking account of
calibration parameters which were determined during measurement
operation and/or calibration parameters which were determined and
stored during an earlier calibration.
[0117] The calibration method or evaluation method is performed
advantageously, at least in part, by a computer program product
stored on a machine-readable carrier. By way of example, this can
be done by means of a microcontroller, digital signal processor,
FPGA or ASIC, wherein the corresponding program code can be
provided for example by a device-internal memory, by external
storage media or by means of electromagnetic waves.
[0118] FIG. 2 illustrates in schematic form a further embodiment of
an angle measuring system according to the invention. It symbolizes
one of the many alternative embodiments comprising a plurality of
reading heads 1a, 1b, 1c, 1d, which can be fitted in a wide variety
of positions with respect to one another. In this case, the
calibration method can be performed analogously to the sequence
described with respect to FIG. 1.
[0119] By way of example, the angle measuring device illustrated in
this figure can, by virtue of a double 180.degree. arrangement of
the reading heads 1a and 1b, and respectively 1c and 1d, be
determined particularly the 1st harmonic repeatedly and with full
resolution. As a result of the 45.degree. degrees arrangement 4 of
the pairs among one another, it is furthermore possible
specifically to determine the 4th harmonic with full resolution. As
a result of the permutation already mentioned, by way of example an
angle of 225.degree. can also be evaluated. As a result of the
double arrangement in pairs, a double evaluation is effected and,
consequently, a further increase in the accuracy of the measurement
value can be obtained by using a compensatory calculation, for
example averaging. Here, too, the angle of 45.degree. is only one
possibility chosen by way of example from any desired angles.
[0120] FIG. 3a shows a further embodiment with three reading heads
1a, 1b, 1c, in which case, however, only two reading heads 1a and
1b are arranged as an opposite pair and the remaining reading head
1c is arranged in an angular position in which it is possible to
determine as many harmonic errors as possible of this measuring
system with a significant proportion. By means of such an
arrangement, the dominant errors of the measuring system which are
to be expected in a structurally governed fashion can be determined
with high accuracy.
[0121] FIG. 3b shows a further arrangement of four reading heads
1a, 1b, 1c, 1d as an example of a further embodiment of an angle
sensor according to the invention. As indicated schematically by
the dash-dotted lines as angular positions of the reading heads,
this arrangement of the reading heads 1a, 1b involves two opposite
reading heads which detect the position code at two angle positions
offset by 180.degree.. An accurate detection of the first harmonic
is thus achieved. In the case of this arrangement, two further
reading heads 1c, 1d in further angular positions achieve an
accurate detection also of those harmonics which cannot be detected
by the 180.degree. arrangement. In this case, it is possible to
determine the angle errors respectively with all possible
combinations of two of the reading heads 1a, 1b, 1c, 1d and,
consequently, for example by means of a compensatory calculation,
it is possible to combine the coefficients that were calculated
with 6 angular positions spaced apart differently.
[0122] FIG. 4 illustrates a flowchart of the calibration method by
the blocks 21 to 24 and of the angle measuring method by the blocks
31 to 34.
[0123] The illustration shows the calibration method according to
the invention that can be carried out without a reference system
for an angle measuring device 7 comprising a code carrier 2
carrying an absolute position code, and at least two reading heads
1a, 1b, 1c, 1d which have a fixed, known angular position 4 having
a different angular distance, in particular of more than 50
degrees, for example of about 150 degrees, relative to one another
and in each case detect the position code at least partially, such
that an absolute angular position value of the respective reading
head 1a, 1b, 1c, 1d relative to the code carrier 4 can be
determined.
[0124] The code carrier 4 is rotatable relative to the reading
heads 1a, 1b, 1c, 1d and different angular positions 3 of the code
carrier 4 relative to the reading heads 1a, 1b, 1c, 1d can thus be
assumed.
[0125] By means of the following steps: [0126] determining the
angle position values 21 of the reading heads 1a, 1b, 1c, 1d in one
angular position, [0127] determining an angle error 22 by comparing
the difference between the angle position values of the reading
heads with the known angular position 4 of the reading heads 1a,
1b, 1c, 1d with respect to one another, [0128] repeating the steps
of determining 23 the angle position values and the angle error for
a multiplicity of varying angular positions, and [0129] carrying
out a mathematical evaluation method 24 comprising [0130]
determining the parameters of a mathematical function quantifying
the angle errors, [0131] determining calibration parameters 40 as
the parameters of the quantifying mathematical function or as a
correction or code table which is derived from the parameters. the
calibration method is carried out.
[0132] The associated angle measuring method for determining an
angle measurement value with the aid of the angle measuring device
7, which comprises a code carrier 2 carrying an absolute position
code, and at least two reading heads 1a, 1b, 1c, 1d which have a
fixed, known angular position 4 having different angular distances,
in particular of more than 50 degrees, for example of 70 to 180
degrees, preferably of 140 to 170 degrees, for example of 150
degrees, relative to one another and in each case detect the
position code at least partially, such that an absolute angular
position value of the respective reading head relative to the code
carrier can be determined.
[0133] The code carrier 2 is rotatable relative to the reading
heads and different angular positions of the code carrier relative
to the reading heads can thus be assumed.
[0134] By means of the following steps: [0135] determining an angle
position value 31 by means of at least one of the reading heads
which detects at least one part of an absolute position code of a
code carrier; [0136] determining the angle measurement value 32 by
correcting the angle position value on the basis of calibration
parameters 40 determined in the context of the calibration method;
the angle measurement value 33 is provided, and this sequence is
repeated 34.
[0137] The block 21 represents the step of determining the angle
position values of the reading heads in an angular position.
[0138] This is followed by--symbolized by the block 22--determining
an angle error by comparing the difference between the angle
position values of the reading heads with the known angular
position of the reading heads with respect to one another.
[0139] The arrow 23 symbolizes the step of repeating the process of
determining the angle position values and the angle error of a
multiplicity of varying angular positions, which is carried out, in
particular, until repeating the steps of determining the angular
position values and the angle error takes place in a measurement
range of the angle measuring system, which measurement range is
subdivided into subranges, a number of times such that specific
subranges are covered by the varying angular positions with a
defined density, in particular wherein the angular positions are
distributed approximately uniformly over said specific
subranges.
[0140] Block 24 symbolizes carrying out a mathematical evaluation
method by expanding the angle errors determined for the respective
angular positions in a mathematical series and determining
calibration parameters as a set of coefficients of the mathematical
series or as a correction or code table which is derived from the
set of coefficients, and is represented by block 40.
[0141] The first step of the evaluation method is symbolized by
block 31 and shows the step of determining an angle position value
by means of at least one of the reading heads which detects at
least one part of an absolute position code of a code carrier.
[0142] Block 32 implements the step of determining the angle
measurement value by correcting the angle position value on the
basis of calibration parameters from block 40 determined in the
context of the calibration method.
[0143] Finally, block 33 stands for providing the angle measurement
value, and the arrow 34 stands for possible repetition of the angle
measurement.
[0144] FIG. 5 illustrates a flowchart of the calibration method,
representing the sequence of a continuous calibration in parallel
with measurement operation.
[0145] The angle position values which are detected in the step
represented by block 31 are in this case first corrected in block
32 on the basis of the calibration parameters from block 40 and are
then provided for further processing in block 33.
[0146] In parallel with the correction of the angle position
values, on the basis thereof in block 22 the determination of the
calibration parameters from block 40 is performed or the values
thereof are increased by continuous adaptation in terms of their
accuracy. The arrow 34 symbolizes the repeated iteration of the
sequence during measurement operation.
[0147] In this case, the values of the calibration parameters can
optionally be fed to a further evaluation (represented by block
50), which, for example in the case of abrupt changes in values,
can initiate a warning or determine an estimated value for the
accuracy of the angle measuring device from these values.
[0148] The calibration parameters can also optionally be stored in
a memory 51, for example in order already to have start values upon
switch-on, to be able to identify a misalignment by comparison of
the determined and stored values, or the like.
[0149] The two optional functions, represented by 50 and 51,
respectively, can analogously also be included in the sequence
illustrated in FIG. 4.
[0150] FIG. 6a shows by way of example a use of angle measuring
systems 7 according to the invention, symbolized by their rotation
axes and an arrow for representing the rotatability, in an
articulated arm coordinate measuring device for measurement
tasks.
[0151] FIG. 6b illustrates by way of example a geodetic measuring
device which uses two angle measuring systems 7 with the
calibration method according to the invention. This device can be
calibrated or recalibrated during or before use in the field or
after an explicit starting of the calibration by single or multiple
traversal of the measurement range or parts thereof.
[0152] FIG. 7 shows by way of example a graphical illustration of a
series expansion of an angle error 10 into a first to fourth
harmonic oscillation 11, 12, 13, 14.
[0153] A schematic illustration of an embodiment of optoelectronic
reading heads that can be used in this method is illustrated by way
of example in FIG. 8. In this case, a projection of a position code
onto an optoelectronic sensor element 9 is specifically
illustrated.
[0154] In this case, the position code is illustrated in the form
of a code disk 2 operated in the transmitted-light method. By way
of example, code detection on the basis of reflection, imaging,
shadow casting, holography, by means of self-luminous encoding, or
other known methods, could likewise be used for this purpose.
[0155] Optical deflection and diffraction by mirrors, lens systems,
concave mirrors or any desired combination of such elements can
also be realized.
[0156] In this case, by way of example, LED, lasers, laser diodes,
incandescent lamps, or other components which emit electromagnetic
radiation can be used as the light source 8. In addition, the
radiation thereof can also be guided via lenses and deflection
systems.
[0157] The position code can be embodied in the form of light-dark
patterns, Moire patterns, barcodes, a two-dimensional code,
etc.
[0158] As encoding, all types of code which permit an absolute
position to be determined are suitable, such as, for example,
maximum sequences, a wide variety of binary encodings, analog
intensity profiles, etc.
[0159] In general, the reading head must be able to make the
position code or parts thereof electronically evaluatable. In this
example, this is done by means of an optical sensor line 9,
realizable for example as an arrangement of photodiodes or as a CCD
chip. Whether the position code is converted into electrical
signals in the form of optical, magnetic, capacitive read-out in
one-dimensional, two-dimensional, three-dimensional form is not of
importance for the calibration method.
* * * * *